Hyderabad, India
Lalit Surisetty.
I build systems that turn complexity into clarity
I work at the intersection of engineering, data, and real-world problems. From backend systems and data pipelines to machine learning and decision tools — I build, ship, and improve systems that make a tangible impact.
Engineering
Scalable systems that run reliably
Data
Pipelines, processing and data products
Intelligence
ML solutions that learn and adapt
Decisions
Tools that drive better outcomes
I have built.I have shipped.I have led.
Numbers that reflect the work, not the words.
0+
dashboards & services shipped
0+
people coordinated
0
ML & engineering systems
0+
conferences delegated
Philosophy
01 / Hyderabad
Everyone ships dashboards. I ship decisions.
A chart is easy to make and easy to distrust. The real work happens before it: building resilient backend services, verifying data pipelines, and stress-testing model assumptions until they either break or hold. What survives is what lands in production.
“Analysis informs. Software systems deliver what happens next.”
doc-unify
An offline-first document intelligence microservice that ingests heterogeneous PDF, image, DOCX, and PPTX files, builds a vector index in Postgres (pgvector), automatically proposes a unified schema across the corpus, and extracts structured records with cell-level provenance and confidence scores.
What I've learned
Models are good at everything except the part that matters.
A model will hand you an answer in a second flat, and it will sound sure of itself whether it's right or not. That's the easy ninety percent — producing a first draft of almost anything, fast. The other ten is deciding whether to believe it, and that part still takes a person who has read the data closely enough to doubt it.
The model
Fast, cheap, confident — and occasionally wrong without knowing it.
The judgment call
Slow, expensive, full of doubt — and right when it actually counts.
How it happens
How I get from a question to something shipped.
Question
Is this data trustworthy? Is the premise even right?
Model
Build the pipeline, the model, the thing that produces an answer.
Validate
Check it against reality. Report what actually happened, not what should have.
Ship
Get it running somewhere real, not just in a notebook.
Experience
Full-stack software engineer working across ML, AI, data systems, risk, cybersecurity, fintech, and SaaS.
FastAPI and React services end to end, distributed ML pipelines, RAG and agentic systems, and FIDIC-contract risk and exposure tracking at CITIC — different domains, same underlying work: turning a messy, high-stakes input into a system someone can actually build a decision on.
CITIC Middle East Contracting L.L.C
Junior Risk Analyst
Dubai · Jun 2025 — Jul 2026
Off-campus internship that converted to full-time, taken during final year — SRM's academic structure allows this.
- Logged ~800 risk-register entries across 5 active projects, covering credit files, transactions, and operational risk tickets, for senior management review.
- Reviewed FIDIC-based clauses on ~80 contracts with the legal team, flagging liability, indemnity, and penalty exposure before contract award.
- Vetted 50 subcontractors and suppliers through financial due diligence, supporting vendor pre-qualification and reducing counterparty risk.
- Tracked AED/USD/RMB exposure on procurement and cross-border payments against defined limits, publishing monthly variance reports for senior management risk reviews.
Innodatatics
BI Intern
Hyderabad · Dec 2024 — Mar 2025
- Built 30 Power BI/Tableau dashboards integrating 8 data sources across business, risk, and ESG data, delivering decision-ready reporting for internal stakeholders.
- Cleaned and transformed data across 8 core business datasets using Python, SQL, Pandas, and NumPy, powering financial, sustainability, and risk analytics.
- Built the ESG Risk & Sustainability Intelligence Platform, integrating 20–30 core material metrics across green-building and carbon data, aligned with major ESG rating frameworks, to support sustainability-rating improvement.
Skills
The stack behind everything above — languages, ML tooling, and infrastructure I reach for by default.
Primary stack
Beyond the desk
A computer science degree, and the coursework that shaped it.
B.Tech, Computer Science Engineering
SRM University, Chennai
2026
- Data Structures & Algorithms
- Database Management Systems
- Operating Systems
- Computer Networks
- Machine Learning
- Distributed Systems
Leadership
Associate Director, SRM MUN Society
May 2024 — Jul 2025
Best Delegate, more than fifty times over, across conferences on three continents' worth of MUN circuits.
Certifications
Supervised Machine Learning
Stanford University
Advanced Learning Algorithms
Stanford University
never finished, only shipped